Figure 1.

Sentences were presented one word at a time using Rapid Serial Visual Presentation. The duration of each non-target word was 300 ms. Target words were marked in red (targets are also bold here) and remained on the screen until they were spoken by the participant.